That's 17 million HOUSEHOLDS that struggled to provide food and 6.8 million HOUSEHOLDS that had to skip meals. I wouldn't call that rare.
[1] https://www.ers.usda.gov/publications/pub-details/?pubid=107...
People say this as if we don't have direct statistics about how hungry poor americans are.
Maybe your ASSERTION that poor people should have enough food because "it's cheap" isn't right and you should investigate that.
"Some folks did a study with 900 people. They found the same correlation that the original study did. But when they controlled for household income, they found most of the correlation disappeared."
So original study didn't control for income. If the original study claimed it across all incomes, and then it mostly went away when others controlled for income, then the delayed gratification strong correlation wasn't really for all incomes, right?